Abstract
For anyAC 0 functionf ofn bits, there is a polynomialp such that anyp(logn)-wise decomposable distribution “fools”f. In other words,f cannot distinguish between the pseudorandom strings in the distribution and truly random strings. The polynomialp depends only on the size and depth of the circuit computingf.
This subsumes and extends the class of distributions that were previously known to foolAC 0 functions, and partially answers an open question posed by Linial and Nisan in 1990, as to whether every polylog-wise independent distribution foolsAC 0 functions or not.
Each polylog-wise decomposable distribution serves as a fixed training set of examples for learning (approximately interpolating) allAC 0 functions computed by circuits of some fixed depth and size. Furthermore, small, natural distributions (training sets) exist that yield deterministic learning algorithms that run in timeO (2polylogn) forAC 0 functions, where the degree of the polylog depends on the size and depth of the circuit to be learnt.
This improves on the randomized algorithms with the same time complexity given, for example, by Linialet al. in 1989, where the examples for the training set are picked randomly from specific distributions.
Similar content being viewed by others
References
M. Bellare, A technique for upper bounding the spectral norm with applications to learning. InProc. 5th Ann. IEEE Symp. on Computational Learning Theory (COLT), 1992, 62–70.
J. Bruck, Harmonic analysis of polynomial threshold functions.SIAM Journal of Discrete Mathematics,3:2 (1990), 168–177.
R. Boppana and M. Sipser,The complexity of finite functions. Technical Report MIT/LCS/TM-405, Massachussetts Institute of Technology, Laboratory for Computer Sciences, 1989.
J. Bruck andR. Smolensky, Polynomial Threshold Functions, AC0 Functions and Spectral Norms.SIAM Journal of Computing,21:1 (1992), 33–42.
H. Dym and H.P. McKean,Fourier series and integrals. Probability and Mathematical Statistics series, Academic Press, 1972.
M. Down and M. Sitharam,Shannon bounds for functions over ℤ 2 n . Technical Report #90-12-1, Kent State University, Department of Mathematics and Computer Science, 1990.
M. Furst, J. Jackson, and S. Smith, Improved learning ofAC 0 functions. InProc. 5th Ann. IEEE Symp. on Computational Learning Theory (COLT), 1992, 317–325.
M. Furst, J. Saxe, andM. Sipser, Parity, circuits and the polynomial time hierarchy.Mathematical Systems Theory 17 (1984), 17–27.
J. Håstad,Computational limitations of small depth circuits. Ph.D. thesis, Massachussetts Institute of Technology press, 1986.
J. Kahn, J. Kalai, and N. Linial, The influence of variables on Boolean functions.Proc. 29th Ann. IEEE Symp. Foundations of Computer Science (FOCS), 1988, 68–80.
E. Kushilevitz andY. Mansour, Learning decision trees using the Fourier transform.SIAM Journal of Computing 22:6 (1993), 1331–1348.
N. Linial, Y. Mansour, and N. Nisan, Constant depth circuits, Fourier transforms, and learnability. InProc. 30th Ann. IEEE Symp. on Foundations of Computer Science (FOCS), 1989, 574–579. To appear inJACM.
N. Linial andN. Nisan, Approximate inclusion-exclusion.Combinatorica 10:4 (1990), 349–365.
F.J. MacWiliams and N.J.A. Sloane,The theory of error-correcting codes. North Holland, 1977.
N. Nisan and A.W. Widgerson, Hardness vs. randomness. InProc. 29th Ann. IEEE Symp. on Foundations of Computer Science (FOCS), 1988, 2–11. To appear inJCSS.
N. Nisan and M. Szegedy, On the degree of Boolean functions as real polynomials. InProc. 24th Ann. ACM Symp. on Theory of Computing, 1992, 462–467.
A.A. Razborov, Lower bounds on the monotone complexity of some Boolean functions.Soviet Mathematics Doklady,31 (1985), 354–357.
C.E. Shannon, Communication in the presence of noise. InProceedings of the IEEE,37 (1949), 10–21.
K.I. Siu andJ. Bruck, On the power of threshold circuits with small weights.SIAM Journal of Discrete Mathematics,4:3 (1991), 423–435.
A.C. Yao, Lower bounds by probabilistic arguments. InProc. 24th Ann. IEEE Symp. on Foundations of Computer Science (FOCS), 1983, 420–428.
A.C. Yao, Separating the polynomial time hierarchy by oracles. InProc. 26th Ann. IEEE Symp. on Foundations of Computer Science (FOCS), 1985, 1–10.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Sitharam, M. Pseudorandom generators and learning algorithms forAC 0 . Comput Complexity 5, 248–266 (1995). https://doi.org/10.1007/BF01206321
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF01206321